IMB3-Miner: Mining Induced/Embedded Subtrees by Constraining the Level of Embedding

Abstract

Tree mining has recently attracted a lot of interest in areas such as Bioinformatics, XML mining, Web mining, etc. We are mainly concerned with mining frequent induced and embedded subtrees. While more interesting patterns can be obtained when mining embedded subtrees, unfortunately mining such embedding relationships can be very costly. In this paper, we propose an efficient approach to tackle the complexity of mining embedded subtrees by utilizing a novel Embedding List representation, Tree Model Guided enumeration, and introducing the Level of Embedding constraint. Thus, when it is too costly to mine all frequent embedded subtrees, one can decrease the level of embedding constraint gradually up to 1, from which all the obtained frequent subtrees are induced subtrees. Our experiments with both synthetic and real datasets against two known algorithms for mining induced and embedded subtrees, FREQT and TreeMiner, demonstrate the effectiveness and the efficiency of the technique.

Khai Phá Cây Con Thường Xuyên Trên Cơ Sở Dữ Liệu Weblogs

“…Một số tác giả sử dụng cơ sở dữ liệu cây weblogs CS-LOG [25,26,28]. Các thuật toán của các nhóm tác giả này đều có những điểm mạnh và điểm yếu vì vậy việc so sánh hiệu suất thực hiện chỉ mang tính chất tương đối.…”

BEST: An Efficient Algorithm for Mining Frequent Unordered Embedded Subtrees

“…This enumeration is a bottom-up approach that generates non-redundant candidates [18]. A candidate generation technique can generate valid frequent and infrequent candidates as well as invalid frequent and infrequent candidates.…”

“…From fig 6(b), it can be observed that SLEUTH generated more frequent subtrees in comparison to BEST, as it uses the opportunistic pruning technique which does not fulfil the downward closure lemma and may generate pseudo frequent subtrees [18].…”

SEQUEST: mining frequent subsequences using DMA-Strips

“…A task in data mining is characterized by the type of data, structures and patterns to be found [8,9]. Advancements in data collection technologies have contributed to the high volumes of sales data.…”

“…One way to achieve efficient processing is to develop an intermediate memory construct that helps us to easily process and manipulate something. We showed this in [9] by developing an efficient intermediate memory construct called Embedding List (EL) [9,10] to efficiently mine frequent embedded subtrees. The problem of using such intermediate memory constructs is that it demands some additional memory storage.…”